Image synthesizing apparatus and image synthesizing method

Abstract

An image synthesizing apparatus which generates, from plural images with differing resolutions and lighting conditions, a synthesized image in which the resolution of the original image is improved and the lighting conditions match those of the original image. The image synthesizing apparatus includes: a camera which acquires an original image and plural reference images, the reference images being obtained by capturing the same object shown in the original image under differing lighting conditions; a reference image conversion unit which converts the resolution of the acquired plural reference images to the same resolution as the original image; a coefficient calculation unit which calculates, based on the original image and plural reference images, which now have the same resolution, a conversion coefficient used in performing linear combination on the plural converted reference images and generating an original image; and an image synthesizing unit which generates plural converted reference images by converting the plural reference images to a target resolution, generates a linear-converted image by performing linear combination on the generated plural converted reference images using the conversion coefficient, and outputs, as a synthesized image, the generated linear-converted image or an image in which the adjusted linear-converted image has been adjusted.

Description

CROSS REFERENCE TO RELATED APPLICATION [0001] This is a continuation of PCT application No. PCT / JP2006 / 313608, filed Jul. 7, 2006, and designating the United States of America.BACKGROUND OF THE INVENTION [0002] (1) Field of the Invention [0003] The present invention relates to an image synthesizing apparatus which, having been inputted with plural images of differing spatial resolution (hereafter, referred to simply as “resolution”) resulting from differences in enlargement ratios and so on at the time of capturing the images, generates a synthesized image of a higher resolution, and particularly relates to a technique for improving the resolution of an image using plural images between which exist differences in resolution and lighting conditions. [0004] (2) Description of the Related Art [0005] As an image synthesizing apparatus which synthesizes high-resolution images using conventional image processing technology, there exists an image synthesizing apparatus which, upon being pr...

AI Technical Summary

Benefits of technology

[0011] Accordingly, an object of the present invention, which has been conceived in light of the aforementioned problems with the conventional technology, is to provide an image synthesizing apparatus which generates, from plural images with differing resolutions and lighting conditions, a synthesized image in which the resolution of the original image is improved and the lighting conditions match those of the original image, and an image synthesizing method for achieving the same.

[0013] Accordingly, it is possible to generate a high-resolution synthesized image with lighting conditions identical to those in the original image even in the case where the lighting conditions of the original image and plural reference images differ, by performing linear combination on the reference images using a conversion coefficient. Furthermore, it is possible to calculate the correct conversion coefficient, even when resolutions differ, by calculating the conversion coefficient using pixel values after processing that equalizes the resolution of the original image and the reference images has been performed; as a result, a more highly accurate synthesized image can be generated.

Problems solved by technology

However, there is a problem with the image synthesizing apparatus that generates a high-resolution image using the aforementioned conventional technology in that when lighting conditions such as the position, number, strength, and the like of the light sources differ between the low-resolution original image and the high-resolution reference image, the generated high-resolution synthesized image ends up having lighting conditions different from the original image.

Moreover, there is another problem with the image synthesizing apparatus that generates an image with arbitrary lighting conditions using the aforementioned conventional technology in that when images with differing resolutions are inputted, the correct linearization coefficient cannot be calculated due to the resolution of corresponding pixels being different; therefore, a synthesized image with the correct lighting conditions cannot be generated.

Images